Olivomycin A is one member of the aureolic acid group of antitumor antibiotics. Certain members of the group are used clinically for treatment of testicular tumors, but toxic side effects limit the usefulness of these compounds. We propose to develop a convergent, stereoselective synthesis of olivomycin A, and thereby define a strategy that may be useful for synthesis of analogues with unnatural oligosaccharide residues and improved therapeutic indices. In the preceding grant period we completed syntheses of the aglycone, olivin, and the AB disaccharide. In the new grant period we will focus our efforts on solving the last remaining chemical problem, namely the development of an efficient method for synthesis of beta-2-deoxyglycosidic linkages. This new methodology will then be used in the synthesis of the CDE trisaccharide and in the coupling of the AB and CDE units to the aglycone. We also propose to explore the exo-anomeric effect as a new strategy for stereochemical control in the enantio- and diastereoselective synthesis of polyhydroxylated and amino substituted molecules. Efforts will be made to develop efficient chiral gamma-alkoxyallylmetal reagents and chiral glycolate and glycine enolate equivalents. If successfully developed, these new reagents will be applied to the synthesis of castanospermine and swainsonine, two polyhydroxylated alkaloids that are of intense interest as glycosidase inhibitors.